AI-generated Planning Objections

Local planning authorities across the UK are reporting a sharp rise in AI-generated objections to planning applications and policy consultations. Research published in Planning Resource in February 2026 found that nearly nine in ten councils have received what they believe to be AI-authored representations, with the majority saying the issue is growing.

The responses are longer, more complex, and often cite case law and policy that may not even exist. Officers are spending significantly more time processing them. Some authorities are seeing hundreds of bespoke objections to a single application, where previously they might have received a batch of identical template letters that could be dealt with together.

It's an emerging problem. But the emerging debate is veering towards the potential solutions too quickly.

Two camps, one missing insight

The conversation is settling into two positions. One says councils need better tools to process the growing volume of representations. The other says the answer is to control the format of engagement - structured survey platforms that prevent lengthy, AI-generated submissions from being made in the first place.

There's something important being overlooked in both.

A double standard is emerging that nobody is naming directly. On the input side - residents using AI to write responses - there is growing anxiety. AI-generated representations are seen as less legitimate, something to be detected and managed. On the processing side - officers using AI to analyse those responses - there is enthusiasm. Tools that help handle volume and produce reports faster are broadly welcomed.

But you can't have it both ways. If AI assisting residents to engage more effectively is a threat to the integrity of consultation, then AI assisting officers to process responses should invite the same scrutiny. The concern about authenticity only flows one way.

The use of AI on both sides is inevitable

The reality is that AI is already being used on both sides of consultation, and that isn't going to reverse. The question that matters isn't whether AI is involved - it's whether it's transparent, traceable, and helping everyone get to the substance.

On the input side, if someone finally understands a complex planning proposal because AI helped them cut through technical jargon, that's democratisation. Planning is notoriously difficult for the public to engage with. If AI lowers that barrier, that should be welcomed - not treated as contamination.

On the processing side, if AI surfaces the material issues across thousands of responses so the authority can demonstrate every representation was properly considered, that's capacity. It frees officers to apply professional judgement where it's needed most, rather than spending months on manual categorisation.

Both are legitimate uses. The standard should be the same on both sides: can you show your working?

"AI is a way to increase everyone's competency and access to really complicated issues in planning. If somebody has finally been able to understand what a planning proposal is about with the help of AI, and they wanted to respond to consultations before but weren't able to, then I think that's brilliant." - Annette Jezierska, CEO, The Future Fox

Why trying to prevent the problem isn't the answer

Some argue that the solution is to move engagement onto structured survey platforms - multiple choice questions, constrained formats, short-form responses - so that AI-generated text never enters the system in the first place.

We know from direct experience that this doesn't work. When we first built our consultation tools, we started with that approach: quantitative, structured, neat data. Every single council customer asked us for free-text boxes. They wanted to hear from people in their own words. Over time, we've come to appreciate they were right. Free text is more human. Voice capture - which platforms like Commonplace now support - even more so.

And there's a practical problem. People's views are like water - they find the path of least resistance. One recent consultation had thousands of respondents, and half submitted their responses by email or letter because they couldn't be bothered with the structured third-party platform. The council ended up with the worst of both worlds: a platform capturing half the picture, and an inbox full of unstructured text that still needed processing manually.

Constraining how people engage doesn't remove the problem. It just means you lose sight of half of it.

The real question: Can you show your working?

The debate shouldn't be about how to stop people engaging with AI. It should be about how to make sure the planning system can handle whatever comes in - openly, defensibly, and without losing the substance.

.

That means applying a consistent standard to AI use on both sides of consultation. Not anxiety on one side and enthusiasm on the other. The same test: is the process transparent? Is it traceable? Can you demonstrate how every response was considered and how conclusions were reached?

How ConsultAI addresses this

ConsultAI is built to that standard. It processes consultation responses from any source - platform exports, emails, letters, even scanned handwritten submissions - and produces structured, fully editable reports traceable back to every individual comment.

Unlike generic AI tools, ConsultAI is built specifically for statutory and planning consultations. It identifies material planning considerations, categorises by theme, summarises at scale, and produces reports in the format officers and inspectors expect.

Critically, ConsultAI's outputs have already been tested under statutory scrutiny. Winchester City Council's Local Plan - analysed using ConsultAI - was examined by the Planning Inspectorate and found Sound in 2025. No other AI consultation tool has passed that test.

"We would not have met our reporting deadlines without ConsultAI" - Sevenoaks District Council

Ready to transform your team's workload?

The standard is the same one we'd apply to any side of this debate: can you show your working? ConsultAI can.

To find out more and see how to transform your team's consultation workload, get in touch.